Brewing Materials and Processes: A Practical Approach to Beer Excellence

Brewing Materials and Processes

A Practical Approach to Beer Excellence

Editor

Charles W. Bamforth

Department of Food Science and Technology, University of California Davis, Davis, California, USA

Table of Contents

Cover image

Title page

Copyright

List of Contributors

Preface

Chapter 1. Malts

Introduction

What Should a Malt Analysis Tell You?

Analytical Variance

Laboratory Analysis and Brewery Performance

Prevalence of Food Safety Tests—Does It Really Reduce Risk?

Sustainability as a Metric for Malt Supply

Flavor Specification—Useful, a Distraction, or Measure of the Future?

Malt Sensory Analysis Can Improve Selection and Specification of Specialty Malt

Malt Color—Surely One of the Easiest Parameters to Specify?

Is Not Color Much Easier to Determine by Eye?

Summary

Chapter 2. Adjuncts

Introduction

Functions in Brewing

Use of Liquid Adjuncts

Malted Cereals

Wheat Malts

Rye

Sorghum

Summary

Chapter 3. Hops

Introduction

Impact of Hops on Beer Quality

Variability

How Best to Manage Hops to Achieve Beer Excellence

Product Shelf Life

Chapter 4. Yeast

Introduction

Yeast Life Span

Wort Carbohydrates and Yeast

High-Gravity Brewing

Oxygen Demand and Quality Checks

Wort Nitrogen and Yeast

Wort Minerals and Yeast

Final Beer pH

Washing the Yeast

Differentiating Between Plant Ale and Lager Yeast Strains (and Wild Yeast)

Detection of Wild Yeast

Brettanomyces and Specialty Beers

Killer Yeast Strains

Phenolic Off-Flavor

Flocculation and Yeast Cropping

Yeast Cropping From A Vertical Fermentor

Glycogen

Trehalose

Pitching Rate and Consistency

The Master Culture—Cryogenic Storage

Yeast Propagation From Slope (Laboratory Stage)

Knowing Your Yeast Strain's Background

Purchasing Yeast

Yeast Shipping From Central Facilities

Viability and Vitality

Yeast Autolysis

Tracking the History of the Yeast

Summary

Chapter 5. Water

Introduction

Practical Requirements for Water Treatment

Maintenance

Chapter 6. Wort and Wort Quality Parameters

Extract

Final Attenuation

Nitrogen

Thiobarbituric Acid Index

High Molecular Weight β-Glucan

Turbidity

Dimethyl Sulfide/S-Methyl Methionine

pH

Color

Bitter Units

Viscosity

Zinc

Chapter 7. Alcohol and Its Measurement

On Alcohol, Alcoholic Strength, and Measurements

The Properties of Alcohol and Their Measurement

Precision and Accuracy

Outline of Methods to Measure Alcohol

Coupled Oscillating Density Meters and NIR Alcohol Meters

Fermentation and Alcohol Production: The Mass Balance of Fermentation, Carl Balling, and the Theoretical Basis for Brewing Parametrics

Density Meter/NIR Analyzer Calculations and the Tabarie Relationship—An Important Interlude

Measuring Alcohol Content in the Modern Brewery and Brewery Laboratory: Tables, Algorithms, and Software

Introduction to an Online Calculator and Brewers' Calculations Software

Summary

Chapter 8. Flavorsome Components of Beer

Approaches to Measuring Substances Impacting Beer Flavor

Bitterness

Sweetness

Sourness

Saltiness

Other Inorganic Ions

Vicinal Diketones (VDKs)

Dimethyl Sulfide

Methods for Measuring Other Volatile Materials in Beer

Flavor Stability Assessment

Chapter 9. Dissolved Gases

Theory: Laws Governing Gases Dissolved in Liquids

Units: Dissolved Gas Nomenclature

Why Measure Dissolved Gases?

Measurement Technologies—O2, CO2, and N2

Measuring Dissolved Gases: Challenges

Chapter 10. Controlling Beer Foam and Gushing

Introduction

Quality Assurance Methods and Troubleshooting

Good Manufacturing Practices and Brewing Process Optimization

Ingredients

Process Variables and Challenges

Gushing and Overfoaming

Chapter 11. Color

Introduction

Human Color Perception—Three-Dimensional Color Space

Path

Obtaining Visible Beer Color (tristimulus) Data

Beer–Bouguer–Lambert Law

Details of ASTM E−308 Prescription

Beer Color as Part of Quality Control

Summary and Conclusion

Chapter 12. Haze Measurement

The Nature of the Physical Instability of Beer

The Measurement of Haze

Measurement of Bits

Prediction of the Haze Stability of Beer

The Investigation of Beer Hazes

A QA Approach to Minimizing Colloidal Problems in Beer

Chapter 13. Sensory Analysis in the Brewery

Introduction

Information Required by the Brewery From Sensory Analysis of Beer

Evaluation of Raw Materials and Process Aids

Evaluation of In-process Samples

Evaluation of Packaging Materials

Evaluation of Packaged Beer

Product Management

Tasters

Practical Aspects of Beer Tasting

Chapter 14. Microbiology

Introduction

Brewing Process

Product

Summary

Appendix. Practicalities of Achieving Quality

Index

Copyright

Academic Press is an imprint of Elsevier

125 London Wall, London EC2Y 5AS, UK

525 B Street, Suite 1800, San Diego, CA 92101-4495, USA

50 Hampshire Street, 5th Floor, Cambridge, MA 02139, USA

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

Copyright © 2016 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

ISBN: 978-0-12-799954-8

For information on all Academic Press publications visit our website at https://www.elsevier.com/

Publisher: Nikki Levy

Acquisition Editor: Nancy Maragioglio

Editorial Project Manager: Billie Jean Fernandez

Production Project Manager: Lisa Jones

Designer: Matthew Limbert

Typeset by TNQ Books and Journals

www.tnq.co.in

List of Contributors

C.W. Bamforth,     University of California, Davis, California, CA, United States

C.S. Benedict†

R. Biurrun,     Research Institute for Beer and Beverage Production (FIBGP), Research and Teaching Institute for Brewing in Berlin (VLB e.V.), Berlin, Germany

N. Davies,     Muntons, Stowmarket, United Kingdom

A.J. de Lange,     Formerly McLean, Virginia, United States

M. Eumann,     EUWA Water Treatment Plants, Gärtringen, Germany

A.E. Hill,     Heriot-Watt University, Edinburgh, Scotland, United Kingdom

L.T. Lusk,     Formerly Foam and Flavor Development Scientist, Miller Coors LLC

B. Meyer,     Research Institute for Beer and Beverage Production (FIBGP), Research and Teaching Institute for Brewing in Berlin (VLB e.V.), Berlin, Germany

R. Pahl,     Research Institute for Beer and Beverage Production (FIBGP), Research and Teaching Institute for Brewing in Berlin (VLB e.V.), Berlin, Germany

T.R. Roberts,     Steiner Hops Ltd.

I. Russell,     Heriot-Watt University, Edinburgh, Scotland

C. Schaeberle,     EUWA Water Treatment Plants, Gärtringen, Germany

W.J. Simpson,     Cara Technology Limited, Leatherhead, Surrey, United Kingdom

G. Spedding,     Brewing and Distilling Analytical Services, LLC, Lexington, KY, United States

G. Stewart,     Heriot Watt University, Edinburgh, Scotland

† Sadly, our good friend Chaz Benedict died on March 17th 2015.

Preface

I am frequently asked my opinion on the essentials for success for an aspiring new brewing company. I first suggest that they need to be able to get their beer into the marketplace at a price point that customers are prepared to meet but which will allow a satisfactory margin. This is not a straightforward thing by any means, noting that (in the United States) distributorships are still very much influenced by the very large brewing companies and there is only so much space on a supermarket or liquor store shelf. There are literally hundreds of brewing companies at any one time determined to join the thousands already in this increasingly competitive market. Unless they have their own location to serve their beers, it will unfortunately be a struggle. Moreover, even if they do have their own outlet, chances are that the difference between success and failure is the safety net of a food menu and the availability of nonalcoholic beverages.

I also impress upon them the need to brew enough beer to be able to make a livable income. I am well aware of what I call the femtobrewing mentality of some, namely that the only decent brews are those produced on the tiniest possible scale. Beer, however, remains a low-cost product, and the brutal reality is that you need to sell quite a lot of beer to make a living.

Third—and in my view the most critical requirement of all—is the need to deliver a quality proposition. The blunt reality is that (as for any other commodity) beer drinkers will not tolerate inferior products.

In my many years within the brewing industry (approaching 40), I have had the great good fortune to meet many brewers who are committed to quality, in companies large and small. None has surpassed Ken Grossman, founder and President of the Sierra Nevada Brewing Company. His book, Beyond the Pale (Wiley), should be compulsory reading for any brewer, but most definitely those who want a realistic take on what is needed to succeed. Although there are other brewers at the upper end (in volume terms) of the brewing industry who turn to mass advertising and trash the products of others to hawk their wares, Ken has always known that developing a strong brand based on the simple ethos of product excellence is the route to sustained success, with the beer speaking for itself. He poured his first six batches of beer away, unprepared to hit the market with a product until he believed it was right and that he was able to deliver that same degree of excellence brew after brew after brew.

That is, for me, fundamentally the most important definition of quality: the meeting of customer expectation and the avoidance of unwanted surprises.

Moreover, that perception is very much a personal thing. This is why I refuse to pontificate about which specific brands are good and which are bad. What you like to drink (or eat, or watch, or read, or believe, etc.) is a matter for your personal preference and delectation. My selections are very likely to differ from yours. So for some in the industry to decry long-standing lager brands as being yellow fizzy liquid is to, at a stroke, insult the millions of people who like that style of beer, as well as to display a profound ignorance about the essentials of quality as defined earlier. These beers may not have screaming depths of flavor or the most luscious of foams, but they are astonishingly consistent and pass the test of fitness for purpose. The brewers of these beers are skilled and knowledgeable individuals—if a grievance might be leveled at some of the larger brewing companies, then it should be targeted at the frequently ruthless business and marketing strategies of such entities.

Armed with this definition of quality, then, we must consider the routes to take to achieve the goal of the right product every time—and one that will remain in a condition to delight the consumer, no matter when the product is purchased and consumed within a realistic life span for that particular beer. I suggest that several needs must be fulfilled:

1. You must have a trained and knowledgeable team, one that is motivated and respected. As a former boss of mine insisted, there are two things that matter, the people and the quality, and if you look after the first, they will deliver on the second.

2. The facilities must be capable in terms of design to deliver a quality product when operated correctly.

3. Standard operating procedures must be in place, from raw materials-in to products and coproducts-out that are meaningful, up-to-date, and comprehensive, such that if the proverbial bus ran over the brewer, his or her successor could seamlessly step in.

4. Statistically significant methods must be applied judiciously to monitor the process and product, generating information in a timely fashion to satisfy the needs of the operation.

5. Properly maintained instrumentation must be in place to allow those methods to be pursued reliably.

In the Appendix of Brewing: New Technologies¹, I discussed much of the salient material pertaining to these items (the recommended methods of the American Society of Brewing Chemists and the fundamentals of statistics). Also addressed were the essential principles of Quality Assurance versus Quality Control, including Quality Standards such as the International Organization for Standardization (ISO) series and Hazard Analysis and Critical Control Points (HACCP). An updated version of that Appendix is included in this new book.

In the first book, Brewing: New Technologies, we addressed sequentially the issue of quality classified according to the main attributes of foam, flavor, stability (flavor, colloidal, and microbiological), color, gushing, and impact on health. In this second book, I asked the authors to approach the matter in a somewhat different but complementary manner, with a focus directly on practical aspects. The aim is to address in detail the measurement tools and approaches in the context of the nature and significance of the specifications that are applied for a given raw material, process stage, or the final product. The reader will judge if we have succeeded.

Distinguished Professor Charles W. Bamforth, PhD, DSc (Hull), DSc (Heriot-Watt), CSci, FIBD, FRSB, FIAFoST,     California, United States

¹ Bamforth, C.W., 2006. Brewing: New Technologies, first ed. Woodhead Publishing, Cambridge, UK.

Chapter 1

Malts

N. Davies     Muntons, Stowmarket, United Kingdom

Abstract

The potential pitfalls of relying on a current malt specification to predict brewhouse performance have led to the development of many variants of the existing methods or wholly new tests; yet these have largely failed to become mainstream analyses. There is still significant confusion as to the differences in an analysis made using an Institute of Brewing or a European Brewery Convention (EBC) Congress mash. The old-style specification persists and it is essential that the user has a full understanding of what contributes to each analytical parameter, how they interact, the significance of analytical variance, and the linkage of genetics and the environment to the performance of malted barley. Malt specifications are being expanded to include metrics of sustainability requiring detailed carbon footprint of supply chain and process. Flavor and color assessment is now much more refined and can be exploited for product and process control and new product development.

Keywords

Analysis; Brew house performance; Carbon footprint analysis; Color; Flavor; Food safety; Innovation; Malt; Sustainability

Introduction

Love it or loathe it, the traditional malt specification that has been with us for many years is likely to continue in use for the foreseeable future. Notwithstanding the universal acceptance that it is in parts flawed and difficult to interpret, it is still the mainstay of evaluating the suitability of malt for brewing. Over the years many suggestions for novel methods have been researched and suggested as either descriptors of malt quality or indicators of processability. So far, novel methods have had limited acceptance even when these have been extensively peer reviewed. Even with the new methods it can still be difficult to relate the results to brewhouse performance and cause problems when integrating them into a laboratory set up to follow recommended methods which are generally robust and which use recognized equipment and techniques rather than some of the intricacies involved in some newer assays. Familiarity and tradition have bolstered the conventional malt specification and will undoubtedly ensure it remains for some time yet.

What we have to contend with is that the concept of quality is often more perceptive than well defined. In other words, the judgment of the user based on his experience is not always easily and unequivocally traceable back to a specific analytical parameter (analyte). In many instances, one brewer or distiller describes a batch of malt as superb because it runs through his plant like rocket fuel, whereas another brewer finds the very same batch wholly unacceptable. The dilemma is how to resolve the root cause. Is it malt derived or process derived? Whatever the cause, a problem exists, but resolution may lie not with a fault in supply but rather in what has contributed to arriving at a material apparently in specification that could have been achieved in so many ways.

For consumers the link of malt analysis to product quality is not always appreciated. What malt parameters affect the enjoyment of the final product? Many factors have been found to influence drinkability and many could be attributable to malt (Davies, 2006a,b). We are aware of the influence of malt in beer either consciously or subconsciously and that affects our enjoyment of the product (Fig. 1.1). Precisely how biochemistry in malt manufacture affects some of the more curious aspects of the enjoyment of beer is largely unknown and cannot be reflected in the traditional malt specification. For a small handful of those parameters, however, there are clear links to malt. Freshness can be achieved with a malt that is low in factors that contribute to staling. Clarity can be impacted by malt modification and beta glucan level. Foam quality is affected by malt proteins and protein breakdown. Consumer preference is likely to be more influenced by marketing among the public. If a product is described as full of malt flavor and brewed from the finest malt and hops, it may be perceived as such, whereas a product brewed in the same way but advertised with flavors associated with fruits from esters will not be judged malty. The brewing and distilling trade has made a concerted effort to promote beer appreciation and wholesomeness of its raw materials, yet specific malt flavors or attributes still take a back seat in most sensory descriptions. In a slightly tongue in cheek attempt to describe how we enjoy beer, it has even been described mathematically. The controlling variables were beer temperature, number of people drinking, days before being back at work, venue mood, and availability of foods or snack, but with no place for the ingredients to contribute to that enjoyment (Fig. 1.2; Mindlab, 2012).

Figure 1.1  Factors associated with the drinkability of beer. Some factors we can consciously control and some are subconscious based on biochemistry and social or environmental factors. Arguably, the factors shaded in black are influenced by the malt component. Most of those factors can have a positive impact on the drinkability of beer. After Davies, N.L., November 2006a. Malt - a vital part of the brewer's palette, Monograph 34: E.B.C. Symposium, Drinkability, Edinburgh.

What Should a Malt Analysis Tell You?

Of course, this depends who is asking. For the breeder the difference could be as simple as whether the barley is suitable for feed or malting based on broad-brush measures such as hot-water extract, disease resistance, and good agronomy. For maltsters, a good variety would be one which performed well over a wide range of nitrogen (protein) contents and was easy to germinate and kiln. To a brewer often it is extract yield, color potential, clarity, and flavor. An overarching thread in all these requirements is that the best variety will reduce risk for the whole supply chain. Ideally, the breeder needs a variety that lasts up to 10  years to recoup breeding investment costs. Farmers need the certainty that a crop will grow well and be versatile in different seasonal weather conditions. Maltsters need varieties that can guarantee availability of good quality raw material free from disease and requiring the least processing utilities input. Brewers also need a secure supply of raw material and for the malt not to give rise to any negative processing or taint issues while providing good yeast nutrition and contributing to color and/or flavor. The concept of a risk premium to secure good quality malting barley at the right time is something very real in the modern malting supply chain. Risk can, of course, be managed and reduced for raw material availability, quality, and functional aspects of performance.

Figure 1.2  Formula to describe the enjoyment of beer. Nowhere does this equation allow for the product quality or taste. These parameters seem to be left to the brewer to control and are a given backdrop to enjoying the beverage. Although this is not a serious scientific formula, it illustrates an attitude to how alcoholic beverages are judged by the public. From Mindlab, 2012. Formula for the Perfect Pint. www.taylor-walker.co.uk/Media/Documents/PDF/news/WPR-Mindlab-The-Perfect-Pint.pdf.

For many years, breeders have used their extensive knowledge of plant phenotypes to determine those best for selection from crosses which may number 30,000 individual plants. In recent years, they have been aided through marker-assisted selection or genomics. In some cases it has been possible to link genetic sequences with specific barley or malt attributes and then to select or promote the effect of those sequences in new varieties. Markers based on single-nucleotide polymorphisms (SNPs) have rapidly gained popularity among breeders (Mammadov et al., 2012). There are, however, problems with this type of selection. A major hurdle is the highly repetitive nature of the plant genome (Meyers et al., 2001). The aim is to identify genes or quantitative trait loci (QTL) that contribute to a desired trait in the grain, with agronomy, or even in malt manufacture. Rather than generate a plethora of new QTL for scientific interest only, a recently concluded project sought to use the SNP/QTL libraries available around the world to enable breeding of specific desirable traits for malting barley into existing elite United Kingdom breeding varieties (Thomas et al., 2014; Ramsay et al., 2014). It was notable in the early stages of that research work that the major distinctions between barley varieties were associated with the vernalization gene. Unsurprisingly, we have spring and winter barley varieties as the major barley types. This is just one factor that makes selection of new varieties complex. The complex genetic interrelationships in barley make it difficult to relate process performance to such malt specification parameters as beta glucan, Kolbach index (KI), wort viscosity, fine–coarse extract difference, friability, and free-amino nitrogen (Wentz, 2000; Wentz et al., 2004). What is behind such difficulty in identifying how to breed and improve these characteristics? Many of the genetic traits are carried on different parts of the DNA sequence, on different chromosomes, and some genes exert multiple effects (pleiotropy). The result of this is that one gene or a series of linked genes may control a number of parameters on a malt specification, or that indeed some existing malt analytes are effectively redundant. Similar consequences have been found in attempting to relate diastatic power (DP) to improved fermentability (Evans et al., 2009). In that work, it was made clear that the genetic sequences controlling the three major contributing enzymes to DP (β amylase, α amylase, and limit dextrinase) are all carried on different chromosomes along with other currently unspecified Gibberellin-responsive elements. Selection to improve one of those parameters in isolation is effectively not possible and is an underlying reason why so many malting parameters are interlinked such that changing one inevitably affects a number of others. There is a further layer of uncertainty due to the thermal properties of amylolytic enzymes and the sugar profiles created (Duke and Henson, 2009b; Henson and Duke, 2014). Traits are influenced by genetics and the environment interacting which further complicates breeding selection. It is relatively straightforward to introduce a stable change in a barley variety if the trait is controlled by a heritable gene, but very uncertain and unstable when the impact of the environment on that gene is strong (Kavitha et al., 2012). So much that we think we know is due to the malting process on a current malt specification is influenced much further back in the breeding program and upstream in the mashing process.

As long ago as 1914, a paper in the Journal of the Institute of Brewing by Harold and John Heron on the purchase of malt on the basis of analysis lamented we feel sure that the majority of brewers fail to fully appreciate the limitations of malt analysis. Taking long leaps through countless brewing research studies over the next 60+  years, we come to the review of Hyde and Brookes (1978) in which links between Hot Water Extract, Gravity and run-off time were reasonably well correlated. They found that just four main analytes (extract, soluble nitrogen ratio, beta glucanase level, and total nitrogen) could account for 85% of the variability in the brewhouse, extract accounting for the lion's share at 70%. However, they too felt that analysis was too skewed toward commercial transactions stating the information they provide about brewhouse behavior is very limited and that provided about wort or beer quality virtually non-existent. Essentially, a brewer's requirements for malt can be encapsulated as follows: to produce wort as economically as possible that performs well across all brewhouse operations (O'Rourke, 2002).

A traditional malt analysis can be subdivided into five key groups: starch conversion, carbohydrate conversion, carbohydrate extract, color, and enzyme potential. Physical attributes of barley and malt affect all of these, but are not regularly considered important when assessing malt quality. This can, of course, be dangerous. Varieties that differ in grain-size distribution each year will not necessarily mill the same. Quite often, problems with malt processing in breweries can be traced back to an unchanged mill setting. In a year with grains that are wider in diameter (bold) extract yield can be dramatically diminished by generation of dust when mill gap settings are set too small. When size distribution is lower, grains may be only partly milled, giving rise to poorly digestible pieces of endosperm and undigested glucan leading to haze. Mill adjustment is such an important aspect of brewing performance yet is often the source of many hazy beers or reduced brew length. Picking up the problem at the malt specification stage by early discussion in a new season with the maltster could avoid costly volume reduction and filtration time in the brewery.

These days, a malt analysis is very much like a risk management tool. Brewers need consistency of raw materials, yet there will always be seasonal variations and barley malting performance changes throughout the year from harvest onward as the grain matures. Much of that annual change in product is seen only by the maltster who will adjust the process conditions to match the agreed specification. Seasonal changes, in which significant swings arise in the levels of protein or beta glucan or grain-size distribution, can be more of an issue that has to be considered. Grain protein content has been found to correlate with applied nitrogen and beta amylase activity (Yin et al., 2002; Qi et al., 2006). Grain size is known to affect diastatic power and hence can be seasonal (Agu et al., 2007). If the specification requires higher-limit dextrinase activity, for example in distilling or high-adjunct grists, there should be an acknowledgment that this could come at the expense of higher malting loss and reduced extract. That is because this enzyme develops appreciably only after a long germination (Sissons et al., 1993) and it has to be mashed carefully to avoid losses due to heat deactivation during mashing (Stenholm and Home, 1999). A similar trade-off has been found using dehusked barley in which filtration and amylolytic enzyme activity was worse, yet processing and predicted spirit yield improved (Agu et al., 2008). In many ways the malt analysis directs the brewer in how to avoid process difficulties by adjustment of the grist or mashing conditions.

There are many discussion papers on the relevance of laboratory mashes in relationship to brewhouse performance. Are these really an issue? The same is found in small-scale or micromalting in which process conditions that mimic true malting have to be set up quite differently to the main plant to get the best correlation. In the case of micromalt and brewing analyses, the reader is effectively using them as a comparative benchmark. Departures from the norm are just as important as getting an analysis in range.

An issue often arises when improvements in malt analysis are not matched with changes in mash profile. When more-modified malt is specified, it can improve the rate of throughput and the amount of extract and yeast nutrients. However, in many cases the mash profile remains the same as used for malts with higher protein and lower extract levels. The result is that the better malt is overmashed and foam-positive protein can be degraded, extract diminished, and overall produces a poorer product. This illustrates the importance of understanding not just raw-material quality, but the impact of that on subsequent processing.

Analytical Variance

Variance of analyses between reanalysis in the same laboratory of the same sample and different samples by different laboratories is well documented and mathematically explained. This, of course, adds a degree of controversy at times to the adherence of the maltsters to the expected specification. Whether the variation from specification will actually translate into a real effect on brewing is often difficult to prove. However, it remains important for maltsters to understand the potential variance in a method and somehow to estimate the impact of drifting toward the limits of a specification. Statistically, supplying on the edge of a specification band will inevitably result in some analytical tests being outside the specification. Control of this is often agreed between maltster and brewer by setting blending limits for individual batches which are slightly wider than the blended delivery specification. When laboratories enter results on an electronic enterprise resource planning (ERP) system such as SAP it is often not possible to distinguish the significance of the amount that a particular analytical parameter (analyte) is from the desired final specification. Most systems simply allocate a percentage banding around the analyte value. Other systems are available, though mostly bespoke for each maltster, that grade the significance of each individual analyte with respect to brewing performance. This gives a greater nonconforming score to those analytes that have more significance for brewhouse performance than others. Using the bandings for laboratory variance in analysis (r95) for example, beta glucan being out of specification by one r95 band is more likely to have an impact on brewing than moisture. An effective weighting system for the importance of each parameter and the degree of variance from the specification can be determined through many different mathematical algorithms. This type of system has been around in different guises for a number of years under the general heading of quality scoring or quality number (Aalbers and Van Eerde, 1986). An example of such a systematic approach to malt analysis is shown in Fig. 1.3. Each batch of malt produced can be scored for its adherence to specification firstly for each analyte. In this case, a score of 0 is perfection and within specification. Using banding that approximates to the analytical variance of each analyte, a progressively higher score is assigned to results outside the target range. For some values markedly outside the target range, a jump in quality score can be allocated to indicate this is really not an acceptable batch to blend. A combined score for all the analytes can then be calculated and compared against a grading system for how close to specification is the batch overall (Fig. 1.4). It is also then possible to compare malt production quality for a given period. If a specification is reasonably straightforward to produce and the season's barley ideally suited to that malt type, then the majority of malts in the period could be perfect without blending and have a score of 0 (no defects in any parameter). There is then a balance between the number of perfect batches and those falling outside the specification for any reason. Thus one batch that perhaps is dramatically affected by a power outage, for example, does not disproportionately affect the overall plant performance for that period if all other malt batches were well made. This sort of system can also be used to set a target for the adherence of the malting plant to specifications in general by both malt type and by overall performance. In the case illustrated, the quality score for that batch, for just the three parameters selected, is 250 which is just inside band D. A tougher target of band B performance overall could have been set which would equate to just one parameter being outside one quality band in the period. That, of course, may be achievable if only one batch was faulty in the month. For example, if 10 batches of this fictional malt were made in the month and only one was faulty with a score of 250, then the average performance for the month would be 25, which is a band B performance. With any scoring system it is only as good as the parameters being assessed. If there are analyses that potentially could affect brewhouse performance that are not scored, for example LD impact on DP, a key influencer of performance may be ignored because it is not on the specification.

Figure 1.3  Example of a Quality Scoring System for Grading Malt Specifications.

If an analytical result falls outside the specification range, it is assigned a value representing the significance of that to brewing performance. The bands are set approximately matching the analytical variance of the method. Scores can increase more markedly if a parameter is particularly likely to cause a brewing problem over a certain value.

Figure 1.4  Example of Quality Banding System to Evaluate Batch Performance.

Using this system, the maltster could set an overall performance target of band B. This would mean that, on average, no malt batch was more than one band outside the target specification for just one analyte. This would be a very tightly controlled specification. However, the value would most likely be set for all batches produced in a given period. If the majority of those batches were perfectly in specification and just one was out, the overall performance could still be judged satisfactory.

This type of system has also been adopted quite effectively by brewers to evaluate maltsters across a number of important supply-chain performance indicators. Of course, there are merits and demerits of each system and it is often applied as a penalty rather than offering a premium for adherence to very narrow variations in quality. Assigning weighting factors and determining which are the most relevant for maximizing process output entails a large degree of supposition, a reality that is likely to continue. Some comfort can be gained from one single number that represents malt quality or brewhouse performance, but there will likely never be a universally accepted system, and each method will remain an amalgam heavily influenced by mathematical prediction and tradition.

With such a system a judgment has been made as to which parameters are most significant to brewers. It is perfectly possible, however, by understanding the brewer's concerns, to set up quality-scoring systems by customer malt type, not just generically. In this way it builds confidence in the maltster that inevitable batch to batch variations will not have a negative impact on brewing performance.

Laboratory Analysis and Brewery Performance

The limitations of direct correlation between laboratory mash conditions and brewery performance are well documented (Axcell et al., 2001; Davies, 2006a,b; Henson and Duke, 2008). Attempts have been made to harmonize isothermal Institute of Brewing (IoB) and EBC Congress mashes to resolve some of these issues and perhaps more accurately reflect brewhouse yield (Evans, 2010; Evans et al., 2011). Why have the old methods prevailed if they are acknowledged as difficult to interpret? Many years of tradition and use undoubtedly have embedded the interpretation of the methods into the minds of analysts, brewers, distillers, and breeders. The use of the specification is to minimize variation rather than predict brewhouse performance. It is a point of reference to compare different suppliers, different crop years, and new varieties and to match raw materials purchasing to customer requirements. As such, it is a well-established benchmark. Choosing the best of the methods for your particular application is important. For example, in Germany it is now common to use just the fine-grind isothermal grist for varietal comparison because the finer grind is closer to commercial practice and is considered a stricter test of barley performance than a temperature-programmed mash.

Essentially, a laboratory analysis is useful if you know what you want out of a raw material and you understand the implications and practical consequences for your process. As such it can be viewed as a safeguard rather than stifling your choice of the best materials available (Axcell et al., 2001). It has to be remembered that malts with apparently the same analytical specification can have widely different performance in brewing and distilling (Axcell, 1998). Table 1.1 indicates why a large degree of interpretation is required in assessing a malt-analysis sheet. It is very important to understand what the possible causes of variation in an analysis could be rather than assume being in specification is sufficient.

A number of malt analyses effectively result in the same impact on the production process, but it is a matter of personal choice which parameters a brewer or distiller selects to ensure that the process will run smoothly. One of the biggest barriers to applying malt analysis results is the limitations of the laboratory mash which can over or under emphasize what will eventually happen during brewing (Davies, 2006a,b). This interaction of parameters has been reflected in the creation of multiparameter equations to predict performance such as fermentability (Evans et al., 2005, 2010).

Attempts to find newer malt analyses that better reflect process performance have been suggested but still not widely adopted, which have included prediction of wort filtration performance. These include TEPRAL (Moll et al., 1989); Stored Waveform Inverse Fourier Transform (SWIFT) (Stewart et al., 2000; Ford et al., 2001); wort osmolyte concentration (Henson and Duke, 2008; Duke and Henson, 2009a); Fourier Transform Infrared (FTIR) spectroscopy (Titze et al., 2009); and medium-infrared analysis to distinguish between extracts that are similar by conventional methods but differ in sugar profiles (Cozzolino et al., 2014). Single-equipment prediction using Near Infrared Spectroscopy is popular and in regular use for measuring malt protein and moisture at grain intake or breeding-line selection. The accuracy and precision of these tests are excellent provided the predictive equations are based on a sufficiently wide range of appropriate samples for the season and geographical location. The limitation remains that they are simply making a faster prediction of an existing parameter and do not necessarily add to the understanding of that value.

There has also